FAQ: Frequently Asked Questions about ISCCP data/products
ISCCP D1, D2 Surface Pressure Abnormality
Observation: A user has noticed a possible problem with the ISCCP D2 Surface Pressure variable for mountainous latitude longitude coordinates. There seems to be a qualitative change from May 1998 to June 1998. The extremes of pressure are the same (from about 650 mb to 1000 mb), but from June 1998 on there are very few values between 700 mb and 900 mb whereas up to May 1998 there were values spread throughout that range. Note that the ISCCP D1 data is also affected by the above observation.
Answer: The surface pressure reported in D2 comes from the incoming RTOVS data that we process into TV data (atmospheric temperature, humidity etc. profiles used in the radiative model) and that is further used in D1, D2 production. The initial atmospheric profiles from incoming RTOVS are checked and processed as described on page 106 in the attached document. Starting with June 1998 there is a change in the incoming RTOVS data. This change is in surface pressure values that increased by about 20 mb on average. This will explain the jump in surface pressure values seen by the user. Also there are some other spurious differences in the coverage and data availability.
For the mountainous region at 18.8S, 68.6W the constancy of the values from June to December (and more or less until the current date) is due to the supplemental climatologies we use to fill in the missing input data in the ISCCP TV product and which in this case is the climatology value of 800mb.
These climatologies are used only in the event that no daily data is available for that cell and no monthly mean could be calculated and used as a fill. (At least 5 values should be present in a month to calculate a monthly mean for that cell and use it to fill for the rest of the missing daily values for that month).
Refer to the ISCCP New Cloud Data Documentation for more on supplemental ISCCP climatologies (page 98, last paragraph) and the fill-in method (page 105).